From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757817AbcH2PY1 (ORCPT ); Mon, 29 Aug 2016 11:24:27 -0400 Received: from mx1.redhat.com ([209.132.183.28]:57146 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757692AbcH2PY0 (ORCPT ); Mon, 29 Aug 2016 11:24:26 -0400 Message-ID: <1472484262.32433.95.camel@redhat.com> Subject: Re: [PATCH RFC v3] x86,mm,sched: make lazy TLB mode even lazier From: Rik van Riel To: Ingo Molnar Cc: "H. Peter Anvin" , serebrin@google.com, peterz@infradead.org, linux-kernel@vger.kernel.org, luto@kernel.org, bp@suse.de, mgorman@suse.de, tglx@linutronix.de, Linus Torvalds Date: Mon, 29 Aug 2016 11:24:22 -0400 In-Reply-To: <20160827080316.GA11325@gmail.com> References: <20160825150515.02c2d8ea@riellap.home.surriel.com> <7E7CF02F-F0B1-493A-98B3-B078174811DA@zytor.com> <20160825170133.0a783ae8@riellap.home.surriel.com> <20160827080316.GA11325@gmail.com> Content-Type: multipart/signed; micalg="pgp-sha256"; protocol="application/pgp-signature"; boundary="=-1ieobx8/fhThNNkDW+o5" Mime-Version: 1.0 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.26]); Mon, 29 Aug 2016 15:24:25 +0000 (UTC) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --=-1ieobx8/fhThNNkDW+o5 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Sat, 2016-08-27 at 10:03 +0200, Ingo Molnar wrote: > * Rik van Riel wrote: >=20 > > On Thu, 25 Aug 2016 12:42:15 -0700 > > "H. Peter Anvin" wrote: > >=20 > > > Why grabbing a lock instead of cmpxchg? > >=20 > > ... and some more cleanups later, this might actually be > > good to merge, assuming it works for Benjamin :) > >=20 > > ---8<--- >=20 > LGTM in principle (it's a pretty clever trick!), just some minor > stylistic nits: Thanks for the review. I have applied the stylistic nits, and turned lazy_tlb_can_skip_flush into a big switch statement as suggested by Linus. > > + =C2=A0*/ > > + for_each_cpu(cpu, cpumask) > > + if (lazy_tlb_can_skip_flush(cpu)) > > + cpumask_clear_cpu(cpu, (struct cpumask > > *)cpumask); >=20 > Please remove the 'const' from the cpumask type definition instead of > this ugly=C2=A0 > cast! I played around with this on Thursday already, and ran out of time to clean that up before going to the next talk. This will be fixed in the next version.=C2=A0 > I'd also like to wait for the Tested-by from Benjamin as well before > we can=C2=A0 > proceeed. Agreed. Ben, a new version is coming up real soon. --=20 All Rights Reversed. --=-1ieobx8/fhThNNkDW+o5 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAABCAAGBQJXxFOmAAoJEM553pKExN6DfaUIAJWBY571Ks7zyxcP0/JHZcJl yIeyjI3L8M+n8sQuXsm63Xt4DHbxzOVLrxpMxu+x6Yd9r1Rx3P0/nRX16OLoikch 3c/gKwA94hTK6G5xf8nmZ8wUS9yRrwY3iqTrwY0Tuy7N0HXWMRoNTqU8hTiYnjcE mkZQN/IjhMk7XZkpE7dhMRzYyYifTTZvUoQDJxsrWPFraWEUziX9c8+co9Sl9rM5 3LTiT9w8n326BNzfQY7aVbNdrcsQ/cKSjaNLk0Lgj3rcTFeK+8Vx46X8B574yQBF 1l+xT/ARCpXwnVb1r1hT+Rqp7csWFUIgvJ0b6dgj2R+t3RYEwqCEO6WvSrBpULI= =9VoM -----END PGP SIGNATURE----- --=-1ieobx8/fhThNNkDW+o5--